基于改进YOLO V5的缺陷识别与定量分析
Defect identification and quantitative analysis method based on improved YOLO V5
摘 要
采用YOLO V5算法对碳纤维复合材料预置夹杂缺陷的识别方法展开研究。为了在提高检测精度的同时保证检测效率,通过添加通道注意力机制、空间注意力机制、使用k-means++重新聚类先验框和优化损失函数等措施改进原算法。利用改进后的网络训练缺陷数据集,每秒处理的图片数量逾12幅,平均精度达到98.8%,召回率为98.1%。与其他算法相比,该算法检测精度和速度都有所提高,可满足实时性和准确性要求。
YOLO V5
通道注意力机制
空间注意力机制
夹杂缺陷
nondestructive detect
YOLO V5
channel attention mechanism
spatial attention mechanism
inclusion defect
Abstract
The YOLO V5 algorithm was used to study the identification method of carbon fiber composite material preset defects. In order to improve the detection accuracy while ensuring the detection efficiency, the original algorithm was improved by adding channel attention mechanism and spatial attention module, using k-means++ to re-cluster the prior box and optimizing the loss function. Using the improved network training defect data set, the number of images processed per second reached more than 12 frames, the average accuracy reached 98. 8%, and the recall rate was 98. 1%. Compared with other algorithms, the detection accuracy and speed of this algorithm have been improved to a certain extent which can meet the real-time and accuracy requirements.
中图分类号 TG115.28 TB332 DOI 10.11973/wsjc202301003
所属栏目 试验研究
基金项目 吉林省教育厅科学技术研究项目(JJKH20210823KJ); 吉林省科技资源开放共享服务平台与科研条件保障项目(20191004022TC)
收稿日期 2022/5/24
修改稿日期
网络出版日期
作者单位点击查看
备注刘洋(1997-),男,硕士研究生,主要研究方向为超声三维可视化
引用该论文: LIU Yang,LI Quanyong,GU Jian,YANG Fan,WANG Wenbo. Defect identification and quantitative analysis method based on improved YOLO V5[J]. Nondestructive Testing, 2023, 45(1): 14～19
刘洋,李全勇,顾健,杨帆,王文博. 基于改进YOLO V5的缺陷识别与定量分析[J]. 无损检测, 2023, 45(1): 14～19
共有人对该论文发表了看法，其中：
人认为该论文很差
人认为该论文较差
人认为该论文一般
人认为该论文较好
人认为该论文很好
参考文献
【1】宋雪旸,张岩,徐成功,等.碳纤维/聚丙烯/聚乳酸增强复合材料的力学性能[J].纺织学报,2021,42(11):84-88.
【2】NELUYB V A,MALYSHEVA G V,KOMAROV I A.New technologies for producing multifunctional reinforced carbon plastics[J].Materials Science Forum,2021,1037:196-202.
【3】苏飞,欧阳晨恺,李枫,等.碳纤维增强树脂复合材料齿槽加工中分层缺陷的形成机制[J].复合材料学报,2021,38(12):4042-4051.
【4】CARRILLO J,RODRÍGUEZ D,VILLAR-SALINAS S.Contribution of CFRP to the shear strength of retrofitted lightly-reinforced concrete panels[J].Journal of Building Engineering,2021,44:102722.
【5】TENG G Y,ZHOU X J,YANG C L,et al.Ultrasonic detection method of micro defects in thick-section CFRP[J].Optics and Precision Engineering,2018,26(12):3108-3117.
【6】刘旭,吴俊伟,何勇,等.基于空耦换能器的碳纤维增强环氧树脂编织复合材料激光超声检测技术[J].复合材料学报,2021,38(9):2822-2831.
【7】张海燕,宋佳昕,任燕,等.碳纤维增强复合材料褶皱缺陷的超声成像[J].物理学报,2021,70(11):165-172.
【8】ZHANG X F,WU X,HE Y Z,et al.CFRP barely visible impact damage inspection based on an ultrasound wave distortion indicator[J].Composites Part B:Engineering,2019,168:152-158.
【9】BOHLOULI R,ROSTAMI B,KEIGHOBADI J.Application of neuro-wavelet algorithm in ultrasonic-phased array nondestructive testing of polyethylene pipelines[J].Journal of Control Science and Engineering,2012,2012:1-9.
【10】RHIM J,LEE S W.A neural network approach for damage detection and identification of structures[J].Computational Mechanics,1995,16(6):437-443.
【11】XU Y,WEI S Y,BAO Y Q,et al.Automatic seismic damage identification of reinforced concrete columns from images by a region-based deep convolutional neural network[J].Structural Control and Health Monitoring,2019,26(3):e2313.
【12】JIANG P Y,ERGU D J,LIU F Y,et al.A review of yolo algorithm developments[J].Procedia Computer Science,2022,199:1066-1073.
【2】NELUYB V A,MALYSHEVA G V,KOMAROV I A.New technologies for producing multifunctional reinforced carbon plastics[J].Materials Science Forum,2021,1037:196-202.
【3】苏飞,欧阳晨恺,李枫,等.碳纤维增强树脂复合材料齿槽加工中分层缺陷的形成机制[J].复合材料学报,2021,38(12):4042-4051.
【4】CARRILLO J,RODRÍGUEZ D,VILLAR-SALINAS S.Contribution of CFRP to the shear strength of retrofitted lightly-reinforced concrete panels[J].Journal of Building Engineering,2021,44:102722.
【5】TENG G Y,ZHOU X J,YANG C L,et al.Ultrasonic detection method of micro defects in thick-section CFRP[J].Optics and Precision Engineering,2018,26(12):3108-3117.
【6】刘旭,吴俊伟,何勇,等.基于空耦换能器的碳纤维增强环氧树脂编织复合材料激光超声检测技术[J].复合材料学报,2021,38(9):2822-2831.
【7】张海燕,宋佳昕,任燕,等.碳纤维增强复合材料褶皱缺陷的超声成像[J].物理学报,2021,70(11):165-172.
【8】ZHANG X F,WU X,HE Y Z,et al.CFRP barely visible impact damage inspection based on an ultrasound wave distortion indicator[J].Composites Part B:Engineering,2019,168:152-158.
【9】BOHLOULI R,ROSTAMI B,KEIGHOBADI J.Application of neuro-wavelet algorithm in ultrasonic-phased array nondestructive testing of polyethylene pipelines[J].Journal of Control Science and Engineering,2012,2012:1-9.
【10】RHIM J,LEE S W.A neural network approach for damage detection and identification of structures[J].Computational Mechanics,1995,16(6):437-443.
【11】XU Y,WEI S Y,BAO Y Q,et al.Automatic seismic damage identification of reinforced concrete columns from images by a region-based deep convolutional neural network[J].Structural Control and Health Monitoring,2019,26(3):e2313.
【12】JIANG P Y,ERGU D J,LIU F Y,et al.A review of yolo algorithm developments[J].Procedia Computer Science,2022,199:1066-1073.
相关信息
